Scientific Article - MRSA and the use of silver dressings

MRSA and the use of silver dressings: overcoming bacterial resistance


Steve Thomas
Surgical Materials Testing Laboratory Bridgend, Wales, UK




  • Introduction
  • Clinical significance of MRSA
  • Cost of MRSA
  • Controlling the spread of MRSA
  • Psychological effects of MRSA infections
  • Significance of MRSA wound infections
  • Treatment of wounds containing MRSA
  • Silver dressings
  • Conclusion
  • References

Published: Nov 2004 
Last updated: Nov 2004 
Revision: 1.1

Keywords: antibiotic resistance; antibacterial agents; antimicrobial agents; methicillin-resistant Staphylococcus aureus (MRSA); silver dressings.

Key Points

1.      Resistant bacteria, particularly methicillin-resistant Staphylococcus aureus (MRSA), are an ever-increasing global threat that has reached epidemic proportions and is extremely costly, both in terms of human suffering and the financial burden it places on healthcare systems.

2.      The battle against bacterial resistance should focus on preventing the spread of contamination by reducing the risk of infection, educating both staff and patients in infection control, and improving standards of hygiene.

3.      The problem of resistant bacteria has been exacerbated by the uncontrolled use of antibiotics, so a range of alternative measures are required to prevent the development of further resistance and manage infections. These include antibacterial preparations and dressings that contain silver.


The spread of antibiotic-resistant strains of micro-organisms such as methicillin-resistant Staphylococcus aureus (MRSA) represents an ever-increasing threat to the health of vulnerable people throughout the world who are obliged to spend extended periods in healthcare facilities. The organism is also responsible for increasing the financial burden placed on such centres and the wider community at large, with the result that precious financial resources are diverted from other areas of need to deal with the consequences of infection.

There is general agreement that the problem of resistance has been exacerbated by the overuse or misuse of antibiotics so, wherever possible, alternative methods are now required to manage topical infections caused by antibiotic-resistant organisms. Open wounds act as an important focus for cross-infection, necessitating the application of appropriate measures to eliminate or prevent the spread of bacteria from such lesions. Some topical products that can be used in the treatment of wound infections are described, with particular emphasis on the potential value of silver-containing dressings.


Staphylococcus aureus is a gram-positive bacterium that exists as a skin commensal in a significant proportion of the population. Despite its ubiquitous nature, it is a recognised potential pathogen [1], with the ability to cause a wide range of infections from localised skin eruptions to life-threatening conditions such as bacteraemia, endocarditis and pneumonia [2]. S. aureus is one of the most common causes of hospital acquired (nosocomial) infections and its pathogenicity is due, at least in part, to the production by the organism of coagulase, an enzyme that clots plasma and thus inhibits host defence mechanisms.

Another important feature of the bacterium is its ability to develop resistance to commonly used antibiotics and antiseptics. The development of resistance to penicillin by S. aureus, mediated by the production of an enzyme called penicillinase, a β-lactamase, was first reported in 1945 [3]. This enzyme is able to break down the β-lactam ring, an important part of the structure of numerous antibiotics, rendering them ineffective. The introduction of new forms of penicillin that are not affected by β-lactamase, such as methicillin, cloxacillin and flucloxacillin, initially provided effective control of clinical infections caused by penicillinase-producing strains of S. aureus. By the early 1960s, however, isolated reports of resistance to these agents began to appear in the literature, even before the antibiotics were used in clinical practice [4]. These resistant strains, which were subsequently isolated from clinical specimens, became known as methicillin-resistant Staphylococcus aureus (MRSA) [5], and now more than 80% of all strains of the organism examined in laboratories are resistant to antibiotics.

The emergence in 1988 of ciprofloxacin-resistant S. aureus in nine hospitals and eight nursing homes in New York City was documented by Budnick and Schaefler [6]. Twenty-four of the 55 patients involved had been treated previously with ciprofloxacin and MRSA was shown to be a contributing factor in at least five of 21 deaths.

The mechanisms by which resistance develops are complex and have been described in detail elsewhere [7],[8], but it has been demonstrated that resistance is associated with the production of a unique, low-affinity penicillin-binding protein (PBP2a) encoded by the chromosomal gene mecA. Neither the protein nor the mecA gene is present in methicillin-sensitive S. aureus (MSSA) [9].

That the development of resistance can be accelerated by inappropriate or excessive use of antibiotics was illustrated graphically by Fukatsu et al, who described how the overuse of third-generation cephalosporins led to a serious outbreak of MRSA in a university teaching hospital [10].

MRSA can also develop resistance to families of antibiotics other than the penicillins. In one reported case, oral rifampicin not only failed to eradicate the organism in four patients who were carrying it but also produced isolates that were highly resistant to the drug [11].

Some strains of MRSA readily spread from patient to patient and are therefore called epidemic methicillin-resistant S. aureus or EMRSA [12]. This term does not relate to the antimicrobial resistance of the organism but rather to its behaviour in an epidemiological context [13]. Like ordinary MRSA, the resistance of EMRSA to other families of antibiotics may vary. In the laboratory, EMRSA are classified according to antibiotic susceptibility, phage reactions and toxin production. These vary in virulence and their propensity to cause infections in specific tissues [2]. Of particular concern are reports of the emergence in Japan, the USA and Eastern Europe of strains with low levels of resistance to vancomycin, currently the treatment of choice for systemic infections due to MRSA [2].

Clinical significance of MRSA

The virulence or pathogenicity of some strains of MRSA may not be very different from those of MSSA [14], but when infections occur they can be more difficult to control.

The risk of mortality associated with a staphylococcal bacteraemia was analysed by Conterno and colleagues [15] after a study was carried out at a general teaching hospital with a high prevalence of MRSA strains in Sao Paulo, Brazil. There was a 39% (53/136) mortality rate within 14 days of developing a bacteraemia and MRSA was isolated from 66% of patients (90/136). Multivariate analysis identified three variables that were significantly and independently associated with mortality: site of entry (lung), shock and resistance of S. aureus to methicillin.

The first epidemics caused by EMRSA occurred in eastern Australia in the late 1970s and this, or a closely related organism, first became a clinical problem in England in 1980-1981, in the north-east Thames region [16]. A survey conducted over a six-month period by the Staphylococcus Reference Laboratory revealed that by 1987-1988 this strain (EMRSA-1) was affecting 50 different hospitals. Eleven other epidemic strains were also identified during the survey, each affecting up to eight hospitals [17]. In one year a single strain of EMRSA, prevalent in south-east England, produced 40 infections, including bacteraemia, pneumonia, surgical wounds, and skin and urinary tract infections, and three attributable deaths in one acute hospital [18].

In 1978, over a six-month period, 61 patients at a university hospital became colonised or infected with MRSA [19]. Most of them (93%) were on surgical wards, and patients with burns became infected more frequently than other acutely ill surgical patients and often remained colonised for 30 days or more. MRSA represents a particular hazard for burns patients and has the potential to cause significant morbidity and mortality in this patient group because disruption of the normal skin barrier and depression of the immune system makes them more vulnerable to colonisation and infection [20].

MRSA can also present a problem in the community, particularly in nursing care and residential homes [21]. In such environments the main risk factors for colonisation and infection include age, underlying conditions, nasal colonisation and the presence of in-dwelling devices such as catheters, tracheostomies and nasogastric tubes [22]. Although there is evidence of in-home transmission [21], MRSA colonisation appears to be more strongly associated with recent hospital admission [17],[23].

A working party report prepared in 1998 by the British Society for Antimicrobial Chemotherapy, the Hospital Infection Society and the Infection Control Nurses Association described a total of 16 strains of EMRSA [2]. In 1996 one strain alone (EMRSA-16) was responsible for 856 reported incidents in 167 hospitals [2]. This particular strain is always resistant to penicillin, erythromycin and ciprofloxacin, but resistance to kanamycin, gentamicin and neomycin is described as 'variable'.

The costs associated with the control of EMRSA-16 can be significant; in 1995 it was estimated that effective control procedures in one hospital saved more than £629,000 in potential expenditure on additional treatments [24].

Over the years the number of centres that have encountered resistant strains has risen sharply and the management of patients with serious infections that do not respond to conventional treatments represents a significant clinical problem. During the period 1991 to 1998, the percentage of S. aureus bacteraemias caused by MRSA rose from 1.5% to 20% in some parts of the United Kingdom, and by the year 2003-2004 it was recorded that of 19,311 reported S. aureus infections, 7,647 (39.6%) were caused by MRSA - an increase of 3.6% on the previous year.

Although many patients become colonised with MRSA, not all develop an infection and many attempts have been made to identify the risk factors that predispose an individual to infection.

Soon after the emergence of MRSA it was recognised that patients in intensive care units were particularly vulnerable to infection. About two-thirds of nosocomial cases and outbreaks have occurred in critical care facilities, and once MRSA has become endemic extraordinary efforts may be required to prevent nosocomial transmission [25].

In one 12-month study, 209 isolates of MRSA were obtained from 39 patients admitted to an ICU [26]. In 23 of these patients, MRSA was the main pathogen, producing either pneumonia, bacteraemia or wound infections. In eight patients death was directly related to MRSA infection. It has been shown that patients with MRSA infections in an acute care facility are more likely to have a prolonged hospital stay, have received prior antibiotics and have severe underlying disease than patients infected with MSSA [22].

In 1997, Coello et al followed a cohort of 479 hospital patients colonised with MRSA [27]. Fifty-three patients (11.1%) subsequently developed 68 MRSA infections. Intensive care setting, the administration of three or more antibiotics, ulcers, surgical wounds, nasogastric or endotracheal tubes, drains, and urinary or intravenous catheterisation were all associated with an increased risk of MRSA infection.

The risk of developing an infection in an intensive care unit was investigated in a large-scale point prevalence survey involving 10,038 patients in 1,417 ICUs in 17 western European countries [28]. On the study day, 21% of patients had ICU-acquired infections, 30% of which were caused by S. aureus. Of these isolates 60% were resistant to methicillin (with a wide intercountry variation). The most commonly reported MRSA infections were pneumonia and lower respiratory tract infections, and the most important risk factor for MRSA was length of stay in the ICU. The presence of MRSA infection also reduced the chance of survival, particularly in association with lower respiratory tract infections, as the risk of mortality was three times higher in patients with MRSA than in those with MSSA.

Cost of MRSA

The cost of managing the problems associated with emerging drug resistance is considerable. According to a report by the National Audit Office, at any given time about 9% of hospital patients have a nosocomial infection, costing the National Health Service as much as £1bn per annum and contributing to the death of an estimated 5,000 people each year [29].

In 1995, the US Office of Technology Assessment reported that antibiotic-resistant infections caused by six species of bacteria in US hospitals cost the country at least $1.3bn (£709m) a year [30].

Controlling the spread of MRSA

Compared with MSSA, infections caused by MRSA are difficult to cure and result in increased mortality [2]. Treatment costs escalate dramatically as a result of increased antibiotic usage and extended hospital stays, with the side effects that result from the use of more toxic antibiotics adding to the cost in human terms. For all these reasons, prevention remains better than cure.

The report of the UK working party on hospital infection [2], and more recently the report by the National Audit Office [31], recommended that despite practical problems, where infection control facilities may be inadequate or in situations where MRSA has become endemic, active intervention to prevent the further spread of the organism is of benefit and should be encouraged.

Hospitals are therefore instructed to devise policies and procedures for dealing with the prevention and control of MRSA based on the advice contained within the report [31]. This includes:

·         Active surveillance and investigation of healthcare associated infection and antimicrobial-resistant organisms

·         Reducing infection risk by controlling the use of invasive devices, instruments and other equipment

·         Reducing reservoirs of infection by improving bed management and isolation facilities

·         Adopting high standards of hygiene and clinical practice

·         Prudent use of antibiotics to minimise the emergence of antibiotic-resistant organisms

·         Improving senior management commitment, local infrastructure and systems

·         Research and development to ensure that technological breakthroughs in prevention and control are rapidly translated into benefits for patients.

It is widely recognised that some bacteria can survive in the environment for extended periods and viable organisms are often isolated from rooms occupied by patients with MRSA. Commonly contaminated objects include the floor, bed linen, patients' gowns and over-bed tables. This means that healthcare professionals can easily transfer bacteria from one patient area to another on inanimate objects such as their clothing, instruments or pens. An outbreak of mupirocin resistance in a dermatology unit was thought to be associated with a contaminated blood pressure cuff and communal shower area [32].

This contamination risk was demonstrated practically in a study in which 65% of nurses, who had performed morning patient-care activities on patients with MRSA in their wounds or urine, contaminated their nursing uniforms or gowns with the bacteria [33]. It also found that MRSA could be cultured from the gloves of 42% of personnel who had no direct contact with patients but had touched contaminated surfaces in the hospital environment.

Berrington and Pedler investigated a domestic ozone generator's ability to decontaminate hospital side rooms previously occupied by patients with MRSA [34] . Measured ozone concentrations reached 0.14ppm. This is enough to cause mild pulmonary toxicity, but although bacterial counts were reduced in the vicinity of the gas generator, the effect elsewhere in the room was limited. MRSA also appeared more resistant to the effects of ozone than MSSA, leading the authors to conclude that the device tested would be inadequate for the decontamination of hospital side rooms.

Effective manual procedures laid down for the cleaning of these areas must therefore be closely followed to eliminate potential sources of infection [35].

Experience suggests that the spread of MRSA can be reduced by attention to simple hygiene procedures. Particularly important is regular handwashing with an effective antibacterial agent such as an iodine-containing soap [36] or a preparation containing triclosan [37],[38]. Aqueous-based hand disinfectants containing chlorhexidine are sometimes used, but are much more effective if they also contain alcohol [39]. The National Audit Report [31] therefore emphasises the need for:

·         Education and training in infection control for all groups of staff, particularly doctors

·         Compliance with guidance on issues such as hand hygiene, catheter care and aseptic technique

·         Controlled antibiotic prescribing in hospitals

·         Improved hospital cleanliness

·         Consultation with the infection control team on wider trust activities such as new building projects.

In addition to inanimate objects, the transmission of viable micro-organisms in the nose and other areas of the body, both by patients and healthy subjects (carriers), must not be overlooked. In healthy individuals, three patterns of carriage can be distinguished over time: about 20% of people are persistent carriers, 60% are intermittent carriers and approximately 20% almost never carry the organism [40].

Casewell described how, in 1959, Williams and colleagues established for the first time that nasal carriers of S. aureus had increased rates of surgical sepsis compared with non-carriers [41]. For half of these patients, the source was the patient's own nose. Subsequent acquisition of tetracycline-resistant strains was associated with even higher rates of infection. The parallels with MRSA are obvious.

Much attention has therefore been focused on the possibility of screening for and then eradicating the organism from the anterior nares of both patients and hospital staff. It has been demonstrated that the acquisition of MRSA by patients may result from exposure to hospital personnel who regularly carry the organism. In one study it was reported that 14 (6%) of 220 staff exposed to patients with MRSA carried the bacterium intranasally, three of whom carried it intermittently for three or more months [19].

The elimination of perioperative nasal carriage of S. aureus can result in a significant reduction in surgical site infections [1]. Both povidone-iodine cream [42] and, more commonly, mupirocin [43] have been shown to be very effective for this purpose, the latter eliminating the organism from over 95% of subjects [44],[45].

In a study to assess the impact of the use of mupirocin ointment on colonisation, transmission and infection with MRSA in a long-term care facility, all 321 residents were monitored from June 1990 to June 1991 [46]. MRSA-colonised patients received mupirocin ointment to nares in the first seven months of the study, and to nares and wounds in the following five months. A total of 65 patients colonised with MRSA received mupirocin ointment which rapidly eliminated MRSA at the sites treated in most patients by the end of one week, but MRSA recurred in 40% of these patients even though they were given weekly preventive treatment. Overall, the MRSA colonisation rate did not change when mupirocin was used in nares only, but was halved when the drug was used in nares and wounds. Mupirocin-resistant MRSA strains were isolated in 10.8% of patients. The authors conclude that although mupirocin ointment is effective in decreasing colonisation with MRSA, its long-term use led to the emergence of mupirocin-resistant MRSA strains. The drug should therefore be reserved for use during outbreaks and should not be used over the long term in facilities with endemic MRSA colonisation.

In a randomised placebo-controlled double-blind trial, Harbarth and colleagues found that nasal mupirocin was only marginally effective in eradicating multi-site MRSA carriage in a setting where MRSA was endemic [47].

Although the incidence of mupirocin resistance is generally low, in one hospital five out of nine MRSA strains isolated in 1992 were resistant to mupirocin, which may signal a serious threat for the future [48]. Changes in mupirocin resistance over an extended period were revealed after surveillance for mupirocin sensitivity in all MRSA strains isolated from colonised or infected patients in a 625-bed public teaching hospital during an epidemic and for three years after it. Mupirocin resistance increased markedly over this period from 2.7% in 1990 to 65% in 1993 due, it is assumed, to the increased use of mupirocin ointment as an adjunct to infection control measures [49].

In Brazil, an epidemic strain of MRSA, which has acquired a novel mupirocin resistance gene, has been reported [50]. After extensive use of topical mupirocin in Rio de Janeiro's university hospital, resistance was observed in 63% of all strains examined. In Uberlandia university hospital, also in Brazil, where mupirocin usage was rare, the resistance rate was 6.1% [51].

Commenting on the development of resistance to mupirocin, Eltringham suggests that any strategy to limit the increase of mupirocin resistance in MRSA should emphasise the importance of controlled antibiotic use of both mupirocin and other agents [52].

The nose is not the only site of carriage of MRSA. A prospective study carried out in the cardiothoracic surgical unit at St Vincent's Hospital in Sydney, Australia, revealed that 27.4% of 84 patients were carrying the bacterium [53]. The perineum was the major site of carriage, with 69.6% of MRSA-positive patients carrying the organism at this site.

Of 67 patients with spinal cord injuries who were carriers of MRSA, the body sites colonised were wounds (58/67), nares (37/67), throat (30/67), urine (27/67) and perineum (17/67) [54].

Although most authors appear to support measures to slow the spread of MRSA, and some have reported a significant reduction in the number of carriers and infections following the implementation of measures to control its spread [53],[55], others remain less convinced. The latter argue that, despite occasional reports of local successes, the steadily increasing prevalence of strains of S. aureus, which are resistant to methicillin, shows that attempts to limit the spread of the bacterium do not work [56]. They argue, somewhat controversially, that efforts to control the spread of methicillin resistance are counterproductive, and that instead energy should be directed towards the control of outbreaks of disease and prevention of the emergence of antibiotic resistance.

Psychological effects of MRSA infections

The potentially serious psychological effects of enforced isolation after a prolonged MRSA infection on patients who may already be depressed as a result of illness or separation from their families are often overlooked. In one study carried out at the National Spinal Injuries Centre at Stoke Mandeville Hospital, Buckinghamshire, 16 MRSA-positive patients with spinal cord injuries aged between 18 and 65 and their matched controls completed a series of questionnaires to measure aspects of the psychological impact on them. The measures used were functional independence, depression, anxiety, and the affective states of anger, vigour, fatigue and confusion. The MRSA-positive patients scored higher in all measures, but only the scores for anger were statistically significant [57].

The problems resulting from the isolation of patients have also been described from a nursing perspective [58]. The author concludes that it is not necessary to isolate all patients who are colonised or infected with MRSA. Provided that basic infection control measures are followed carefully, and that staff, patients and visitors are educated on the risks of MRSA, it is safe to nurse some affected patients on an open ward.

Significance of MRSA wound infections

MRSA, like MSSA, can be isolated from wounds that are apparently healing normally. Although it is unlikely that any attempt would be made to eliminate MSSA in such a situation, a more aggressive approach may be adopted with MRSA because of the potential risk that colonisation may progress to a hard-to-treat clinical infection. Such wounds also represent a serious risk of cross-infection. In one study, inanimate surfaces near affected patients were found to be six times more likely to become contaminated when patients had MRSA in their wound or urine than when the bacteria were present at other sites (36% compared with 6% of surfaces examined) [33].

Although the presence of MRSA will not necessarily lead to the development of a clinical infection, when this occurs it can be fatal. One retrospective study examined outcomes in 204 patients who tested positive for MRSA [59]. Seventy-eight patients were colonised but never developed an infection, 24 were colonised and subsequently became infected, and 102 had one or more nosocomial infections with MRSA at the time of first culture. Of those patients with clinical infection, 25.5% died and it was judged that MRSA had contributed to death in 57.6% of these cases. The authors conclude that it is important to differentiate between infection and colonisation in hospitals where MRSA is endemic so that early, specific treatment may be initiated. It was also suggested that the risk factors for infection should be differentiated from those for acquisition of the organism.

Other authors are less convinced about the threat posed by MRSA. After a comparison of morbidity, the mortality rate and cost factors related to more than 214 documented infections caused by a variety of organisms over a four-year period, Shannon et al conclude that the problems caused by MRSA were relatively minor compared with those caused by Pseudomonas aeruginosa and Escherichia coli [60]. They suggest that the current preoccupation with MRSA infection is unwarranted and unsubstantiated.

Treatment of wounds containing MRSA

Although the presence of MRSA in a wound may influence certain important aspects of a patient's care, it will still be necessary to manage the wound in the most appropriate manner. An open wound colonised or infected with MRSA, or any other potentially pathogenic bacteria, represents a serious potential source of cross-infection. When managing such a wound, it is the duty of the healthcare practitioner to ensure that at all times the treatment provided is in the best interests of the individual concerned and is not likely to create a potential risk to other patients or members of staff.

While elimination of the infective agent is clearly the primary aim, this process may be prolonged so measures must be taken to prevent the spread of contamination during this period. Dressings that readily permit strike-through or shed fibres on removal should be avoided as they may transmit contaminated particles that could easily be carried around the room on air currents, contaminating adjacent surfaces. The use of irrigant solutions to remove adherent dressings may also increase the potential for infected material to be transferred from the wound to the surrounding area, either in droplets that bounce off the wound surface when a jet of solution is applied with force by means of a syringe or even in a gentle trickle that runs down the patient's leg.

Semi-permeable dressings such as films, film-foam combinations and hydrocolloids, which effectively seal off the peri-wound area, may go some way towards addressing these issues as they help to prevent the passage of contaminating organisms both into and out of a wound [61]. However, the use of these products depends on whether their fluid-handling characteristics and performance are appropriate to the condition of the wound and the amount of exudate produced.

If active treatment to eradicate MRSA from a wound is considered appropriate, mupirocin has historically been considered the agent of choice as it has been shown to be effective against MRSA both in laboratory and clinically based studies [1],[62],[63]. For minor skin lesions it is applied in a polyethylene glycol base (Bactroban) but this should not be used on larger wounds because of potential nephrotoxicity caused by the polyethylene glycol. Mupirocin in a paraffin base (Bactroban Nasal) is usually recommended for this purpose but prolonged application should be avoided to reduce the possibility of inducing mupirocin-resistant strains of MRSA [13]. The normal recommended dose is three to four applications a day for five to ten days [64] as once daily application is unlikely to be effective.

Having to apply an ointment to a wound several times a day does not fit well with standard wound management practice, which generally aims to apply a dressing system that may remain undisturbed for several days. It might be argued, however, that because of the potential problem of developing resistance to this antibiotic it should be reserved for use in circumstances in which other dressings or pharmaceutical agents may not be appropriate.

For the management of infected wounds the following antibacterial agents may be considered:

·         Silver sulfadiazine and chlorhexidine (Silvazine), which has been shown to be effective against MRSA in vitro when tested against 50 strains of MRSA [65]

·         Preparations containing povidone-iodine, including Betadine Solution (10% povidone-iodine; PVP-I) and Betadine Cream (5% povidone-iodine) [66],[67]. One study, however, suggests that the antimicrobial activity of Betadine against MRSA was eliminated on dilution to one-quarter strength [68]. These results are at variance with others which showed that povidone-iodine, when diluted 1:100, was bactericidal against both MRSA and MSSA, killing all organisms within 15 seconds [69]

·         Preparations containing cadexomer iodine, which reduced the number of MRSA bacteria in wounds in an animal model, led the authors to conclude that it might be effective in preventing the proliferation of MRSA in wounds [70]

·         Silver sulfadiazine, which performs reasonably well in the laboratory [71], has proved to be less effective than mupirocin both in vitro [72] and in vivo [73]. Yoshida et al[74] applied a cream containing 1% silver sulfadiazine (Geben cream) to 13 patients with skin lesions infected with MRSA and found that the organism was eradicated from only 45.5% of 11 evaluable patients

·         The use of honey for infected wounds is increasing in popularity and a number of dressings or preparations containing it are now available, some of which have been shown to possess good antimicrobial activity against a wide range of pathogenic organisms, including resistant strains [75],[76] .

Other agents that have been recommended for the treatment of MRSA infections include tea tree oil [77] and gentian violet ointment [78], which was claimed, on rather doubtful evidence, to be more effective clinically than povidone-iodine in removing MRSA from pressure ulcers [79]. Extracts of tea are also said to have inhibitory effects on MRSA [80], even at the concentration achieved in infusions prepared for drinking [81].

Ultraviolet light at 254nm has been shown to kill MRSA in vitro [82] and it has been suggested that it might be suitable to treat clinical infections provided that exposure times are appropriate. However, the light is unlikely to penetrate deep enough into the tissues to eradicate the organism, particularly in the presence of slough or necrosis.

In such situations a treatment that combines debridement with effective antimicrobial activity is desirable and the use of sterile maggots of the greenbottle fly Lucilia sericatahas been reported to rapidly cleanse wounds of slough and eliminate infection [83].

Given the problems of finding an effective form of systemic therapy for highly resistant bacteria, the importance of selecting an effective topical treatment cannot be over emphasised. After a study that compared the sensitivities of MRSA isolates collected from 44 consecutive patients against a range of antimicrobial agents, Smoot et al argued that all hospitals should regularly test the materials they use in order to provide the most appropriate form of therapy [84].

One family of dressings that has recently generated a great deal of interest with regard to the treatment of infected wounds depends on the pronounced antimicrobial activity of silver ions. Metallic silver has been used empirically to prevent the growth of micro-organisms since at least the time of Aristotle, who advised Alexander the Great (335 BC) to store his water in silver vessels and to boil it before use. According to White, the preservative properties of silver are still used for this purpose as water tanks on spacecraft are lined with silver to prevent bacterial growth [85]. The remarkable biocidal properties of silver have lead to its inclusion in a wide range of products, including clothing, several types of medical devices and many different types of dressings [86],[87].

Several excellent reviews have been published on the antimicrobial properties of silver and include information on its mechanism of action, development of bacterial resistance, toxicology, clinical indications and the historical background to its use [85],[88],[89],[90],[91].

Silver dressings

Metallic silver is relatively inert but the presence of liquid leads to the release of the silver ion responsible for its biological activity. Silver ions are biocidal at very low concentrations due to the ability of microbial cells to absorb and concentrate silver from very dilute solutions. However, the presence of organic matter significantly diminishes the efficacy of silver[92]. In complex organic biological fluids, concentrations greater than 50ppm [93]and as high as 60.5ppm are needed[94]. Once in the cell the silver binds to and denatures proteins, including DNA and RNA, inhibiting cell replication. Silver in solution exists in three oxidation states: Ag+, Ag++ and Ag+++. Each of these is capable of forming inorganic and organic compounds and complexes, although the Ag++ and Ag+++ forms are unstable or insoluble in water [91]. Ionic silver is active against a wide range of pathogenic organisms but not all forms of silver exhibit antimicrobial activity. Colloidal preparations in which the silver is complexed with albumin or other proteins were once used as antimicrobials, but their use was abandoned as they were shown to be ineffective. Nevertheless, preparations containing colloidal silver are still promoted on the internet and elsewhere as 'health foods', accompanied by extravagant and unsubstantiated claims of their beneficial effects on clinical disorders ranging from influenza to skin conditions such as cuts and warts [95].

In a laboratory-based study, van Hasselt et al [96] examined the antimicrobial properties of three types of colloidal silvers against a range of test organisms, showing a complete lack of activity and concluding that 'claims of its potency are misleading and that there is no place for it as an antiseptic'.

Commercially available silver-containing dressings

A brief review of some of the silver dressings currently available clearly indicates that considerable differences exist between them in terms of their overall structure, and the concentration and formulation of the silver compound responsible for their antimicrobial activity.

Acticoat - Smith and Nephew

Acticoat consists of two layers of a silver-coated, high-density polyethylene mesh, enclosing a single layer of an apertured non-woven rayon and polyester fabric. These three components are ultrasonically welded together to maintain the integrity of the dressing while in use. Silver is applied to the polyethylene mesh by a vapour deposition process which results in the formation of microscopic crystals of metallic silver. Acticoat-7 is similar to the above but consists of two layers of a fine silver-coated mesh enclosing an inner core consisting of two layers of an apertured non-woven fabric made of rayon and polyester. Between the two layers of non-woven fabric is an additional layer of silver coated polyethylene mesh. All five layers are ultrasonically welded together to maintain the integrity of the dressing. Upon activation with water, Acticoat provides a rapid and sustained release of silver ions within the dressing and to the wound bed for three or seven days depending on which dressing is chosen. An absorbent version of Acticoat has recently been introduced in the UK.

Actisorb Silver 220 - Johnson and Johnson

The first silver-containing dressing to make a commercial impact in the UK consists primarily of silver-impregnated activated charcoal cloth.

Arglaes - Unomedical / Medline

Arglaes comprises a mixture of an alginate powder and an inorganic polymer containing ionic silver. In the presence of moisture the alginate absorbs liquid to form a gel and the silver complex breaks down in a controlled fashion, releasing ionic silver into the wound. There are currently two arglaes products available in the UK - a polyurethane film dressing and a postoperative dressing. The alginate powder described above is only available in the US.

Aquacel Ag - ConvaTec

Aquacel consists of a fleece of sodium carboxymethylcellulose fibres containing 1.2% ionic silver. In the presence of exudate, the dressing absorbs liquid to form a gel, binding sodium ions and releasing silver ions.

Calgitrol - Biomedical Technologies Inc

Calgitrol is described by the manufacturer as a silver alginate wound dressing. It consists of an absorbent foam sheet, one surface of which is coated with an alginate matrix containing ionic silver together with a 'cleanser, moisturizer and a superabsorbent starch co-polymer'.

Contreet Foam - Coloplast

Contreet is a polyurethane foam dressing that contains silver, which is released as the foam absorbs exudate.

Contreet Hydrocolloid - Coloplast

The Contreet hydrocolloid dressing, which is based on established standard hydrocolloid technology, also contains a silver complex that is released by wound fluid absorbed by the dressing. This mechanism ensures a sustained release of silver ions as long as the dressing continues to absorb fluid.

Silverlon - Argentum Medical

Silverlon is a knitted fabric dressing that has been silver-plated by means of a proprietary autocatalytic electroless chemical (reduction-oxidation) plating technique. This technique coats the entire surface of each individual fibre from which the dressing is made, resulting in a very large surface area for the release of ionic silver.

SilvaSorb - Medline

SilvaSorb is composed of a synthetic, polyacrylate hydrophilic matrix in which is dispersed or suspended microscopic silver-containing particles. On exposure to moisture the silver is released into the wound in a controlled fashion.

Urgotul SSD - Laboratoires Urgo (Parema)

Urgotul SSD consists of a polyester mesh impregnated with carboxymethylcellulose, white soft paraffin and silver sulfadiazine (SSD).

Silver content of dressings

Typical values for the silver content of each dressing are shown in Table 1

Table 1: Silver content of dressings*

Proprietary Name

Ag content (mg/100cm2)



Calgitrol Ag 




Contreet Foam 


Contreet Hydrocolloid 


Aquacel Ag 




Actisorb Silver 220 


Urgotul SSD (SSD content 3.75%) 


Arglaes powder 


(*Data extracted from previously published studies [86],[87],[91])

The antimicrobial activity of the silver dressing Acticoat was examined by Yin et al, who compared it with silver nitrate, silver sulfadiazine and mafenide acetate using different test systems to determine minimum inhibitory concentration (MIC), minimum bactericidal concentration (MBC) and zone of inhibition against five potentially pathogenic organisms[97]. They showed that although mafenide acetate produced the greatest zone of inhibition, the MBC of the product was much higher than its MIC, indicating that it has a bacteriostatic rather than a bactericidal action. In contrast the MICs and MBCs of the silver-containing products were very similar, indicating that their activity is essentially bactericidal. The authors also showed that although the MIC values for the three silver preparations were very similar when calculated in terms of their silver content, the speed of action of Acticoat was more rapid than the other two products. It was suggested that this is because the metallic silver on the surface of the dressing forms a reservoir of silver ions that are released continuously and, therefore, are always available for bacterial uptake.

The ability of silver to kill antibiotic resistant strains of micro-organisms was investigated by Wright et al in a laboratory-based study during which they compared Acticoat with a solution of silver nitrate and cream containing silver sulfadiazine against 11 antibiotic multi-resistant clinical isolates and showed it to be the most effective at killing all the organisms examined [98]. Based on these results, they suggest that silver products could be used as an alternative to antibiotics to curb the spread of antibiotic-resistant bacteria. Acticoat's ability to kill MRSA and vancomycin-resistant enterococcus (VRE) was also reported by Dunn and Edwards-Jones [92].

When the silver content and antimicrobial properties of 10 silver-containing dressings were compared in a laboratory study, highly significant differences were demonstrated in the activity of the products concerned [87]. Although there was a clear relationship between the silver content and antimicrobial activity of the products examined, the researchers conclude that there are other factors that influence a dressing's ability to kill micro-organisms. These include the distribution of the silver within the dressing (whether it is present as a surface coating or dispersed through the structure), its chemical and physical form (whether it is present in the metallic, bound or ionic state) and the dressing's affinity to moisture, which is a prerequisite for the release of active agents in an aqueous environment. Products that have their silver content concentrated on the surface of the dressing, instead of being bound up within their structure, performed well in these tests, as did dressings that had silver present in the ionic form [87]. When the relevance of one of the test methods used in this study was questioned because one of the products required activation by sodium/chloride ions from wound exudate to exert maximum effect [99], some additional work was carried out on this product and the results were reported separately [100]. These confirmed that the presence of sodium ions can influence the antimicrobial activity of different dressings in different ways, leading the authors to emphasise the need for validated test systems that may be used to conduct meaningful in vitro comparisons of products claimed to possess antimicrobial activity.

Although some bacteria can develop resistance to silver [88], this is not regarded as a serious problem as available evidence suggests that most preparations capable of delivering sustained silver-ion release are effective against MRSA and VRE, and as yet no resistant strains have been encountered clinically [89]. It follows, therefore, that any silver-containing dressing that shows acceptable levels of activity against a range of non-resistant bacterial species should show comparable activity against antibiotic-resistant strains of the same organism.


The use of silver dressings is still in its relative infancy, and as yet the most appropriate use of these materials is not well understood. Although the rapid release of silver may be desirable from a bacteriological point of view, according to White, it is important that the silver present in a dressing is not released into the wound in a short period but slowly over a number of days [85]. This will avoid bolus dosing that could lead to transient elevated tissue, blood and urine levels, minimising the possibility of systemic toxicity.

The author is often asked which one of the current range of silver dressings should be held in a pharmacy or added to a formulary. Unfortunately, there is no easy answer to such questions; a product that may be well suited to one application may be inappropriate for another because of the condition, position or nature of the wound. There are also considerable differences in the cost of the various products which further complicates decision making. From clinical publications and the laboratory studies previously reported Acticoat, for example, is known to rapidly deliver active silver to a wound in concentrations that are likely to exert a marked and rapid effect on an acute infection, something that may be highly desirable in the treatment of a burn or a clinically infected wound. Other studies indicate, however, that the amount of silver released by this dressing is such that it adversely affects cultured keratinocytes in vitro, leading the authors to suggest that the dressing should not be applied as a topical dressing on cultured skin grafts[101]. For this indication it is likely that a dressing that liberates lower concentrations of silver might be more appropriate to prevent the development of infection in an otherwise clean wound. Similarly, a silver-containing product that adheres firmly to the peri-wound skin surrounding a colonised but non-infected wound may be of benefit in preventing the possibility of bacterial transmission while dealing with the underlying infection.

The indications for the use of silver dressings and the choice of specific products depend upon many factors, but the evidence available to date suggests that they have an important role to play in the treatment of infected exuding wounds, including those containing antibiotic-resistant strains of bacteria. However, further work is required to determine how and where each should be used in order to gain maximum benefit. The value of silver-containing dressings remains to be confirmed in the presence of slough and necrosis.

This article is supported by an educational grant from Smith and Nephew. The views expressed in this article are those of the authors and do not necessarily reflect those of Smith and Nephew. Other links: Further information on MRSA can be found at


1. Kluytmans J. Reduction of surgical site infections in major surgery by elimination of nasal carriage of Staphylococcus aureus. J Hosp Infect 1998; 40(Suppl B): S25-29.

2. Duckworth G, Cookson B, Humphreys H, Heathcock R. Revised guidelines for the control of methicillin-resistant Staphylococcus aureus infection in hospitals. British Society for Antimicrobial Chemotherapy, Hospital Infection Society and the Infection Control Nurses Association. J Hosp Infect 1998; 39(4): 253-90.

3. Spink VW, Ferris V. Quantitative action of penicillin inhibitor from penicillin-resistant strain of Staphylococcus. Science 1945; 102: 102-221.

4. Jevons MP. Celbenin-resistant staphylococci. BMJ 1961; 1: 124-25.

5. Griffiths-Jones A. Methicillin-resistant Staphylococcus aureus in wound care. J Wound Care 1995; 4(10): 481-83.

6. Budnick LD, Schaefler S. Ciprofloxacin-resistant methicillin-resistant Staphylococcus aureus in New York health care facilities, 1988. The New York MRSA Study Group. Am J Public Health 1990; 80(7): 810-13.

7. Livermore DM. Mechanisms of resistance to beta-lactam antibiotics. Scand J Infect Dis Suppl 1991; 78: 7-16.

8. Cooper R, Lawrence JC. The role of antimicrobial agents in wound care. J Wound Care 1996; 5(8): 374-80.

9. Katayama Y, Zhang HZ, Chambers HF. PBP 2a mutations producing very-high-level resistance to beta-lactams. Antimicrob Agents Chemother 2004; 48(2): 453-59.

10. Fukatsu K, Saito H, Matsuda T, Ikeda S, Furukawa S, Muto T. Influences of type and duration of antimicrobial prophylaxis on an outbreak of methicillin-resistant Staphylococcus aureus and on the incidence of wound infection. Arch Surg 1997; 132(12): 1320-25.

11. Canawati HN, Tuddenham WJ, Sapico FL, Montgomerie JZ, Aeilts GD. Failure of rifampin to eradicate methicillin-resistant Staphylococcus aureus colonization. Clin Ther 1982;4(6): 526-31.

12. Phillips I. Epidemic potential and pathogenicity in outbreaks of infection with EMRSA and EMREC. J Hosp Infect 1991; 18(Suppl A): 197-201.

13. Hosein IK. Wound infection. 2. MRSA. J Wound Care 1996; 5(8): 388-90.

14. French GL, Cheng AF, Ling JM, Mo P, Donnan S. Hong Kong strains of methicillin-resistant and methicillin-sensitive Staphylococcus aureus have similar virulence. J Hosp Infect 1990; 15(2): 117-25.

15. Conterno LO, Wey SB, Castelo A. Risk factors for mortality in Staphylococcus aureus bacteremia. Infect Control Hosp Epidemiol 1998; 19(1): 32-7.

16. Townsend DE, Ashdown N, Bradley JM, Pearman JW, Grubb WB. "Australian" methicillin-resistant Staphylococcus aureus in a London hospital? Med J Aust 1984; 141(6): 339-40.

17. Duckworth G. Revised guidelines for the control of epidemic methicillin-resistant Staphylococcus aureus. Report of a combined working party of the Hospital Infection Society and British Society for Antimicrobial Chemotherapy. J Hosp Infect 1990; 16(4): 351-77.

18. Law MR, Gill ON, Turner A. Methicillin-resistant Staphylococcus aureus: associated morbidity and effectiveness of control measures. Epidemiol Infect 1988; 101(2): 301-09.

19. Boyce JM, Landry M, Deetz TR, DuPont HL. Epidemiologic studies of an outbreak of nosocomial methicillin-resistant Staphylococcus aureus infections. Infect Control 1981;2(2): 110-16.

20. Cook N. Methicillin-resistant Staphylococcus aureus versus the burn patient. Burns 1998; 24(2): 91-98.

21. Storch GA, Radcliff JL, Meyer PL, Hinrichs JH. Methicillin-resistant Staphylococcus aureus in a nursing home. Infect Control 1987; 8(1): 24-29.

22. Doebbeling BN. The epidemiology of methicillin-resistant Staphylococcus aureus colonisation and infection. J Chemother 1995; 7(Suppl 3): 99-103.

23. Fraise AP, Mitchell K, O'Brien SJ, Oldfield K, Wise R. Methicillin-resistant Staphylococcus aureus (MRSA) in nursing homes in a major UK city: an anonymized point prevalence survey. Epidemiol Infect 1997; 118(1): 1-5.

24. Casewell MW. New threats to the control of methicillin-resistant Staphylococcus aureus. J Hosp Infect 1995; 30(Suppl): 465-71.

25. Boyce JM. Methicillin-resistant Staphylococcus aureus. Detection, epidemiology, and control measures. Infect Dis Clin North Am 1989; 3(4): 901-13.

26. Corticelli AS, Di Nino GF, Gatti M, Melotti RM, Petrini F, Pigna A, et al. Intensive care units as a source of methicillin-resistant Staphylococcus aureus. Microbiologica 1987;10(4): 345-51.

27. Coello R, Glynn JR, Gaspar C, Picazo JJ, Fereres J. Risk factors for developing clinical infection with methicillin-resistant Staphylococcus aureus (MRSA) amongst hospital patients initially only colonized with MRSA. J Hosp Infect 1997; 37(1): 39-46.

28. Ibelings MM, Bruining HA. Methicillin-resistant Staphylococcus aureus: acquisition and risk of death in patients in the intensive care unit. Eur J Surg 1998; 164(6): 411-18.

29. National Audit Office. The management and control of hospital acquired infection in acute NHS trusts in England. London: Stationery Office, 2000; 121 (HC 230 Session 1999-2000).

30. Office of Technology Assessm&l